---
dataset_info:
  features:
  - name: meta
    struct:
    - name: pile_set_name
      dtype: string
  - name: tokenized_text
    struct:
    - name: attention_mask
      sequence: int64
    - name: input_ids
      sequence: int64
  splits:
  - name: validation
    num_bytes: 46389904.317324266
    num_examples: 1825
  download_size: 1557111
  dataset_size: 46389904.317324266
configs:
- config_name: default
  data_files:
  - split: validation
    path: data/train-*
---